Communication deficits are an inherent feature of the progressive cognitive deterioration in Alzheimer's Disease (AD). While mild AD patients frequently deny or minimize their memory problems, they typically complain of their language problems, citing difficulty in following a conversation, losing their train of thought, and difficulty finding the right words to express themselves. This proposal addresses the processes of spoken work recognition as a vehicle for understanding the nature of the communication deficit in AD. It is widely recognized that communication deficits in AD are due in part to a disruption in semantic memory processes. It is generally assumed that phonological and auditory comprehension processes are relatively preserved in AD, although recent data question this assumption (24). Long-term objectives of this project are 1) to explore the relation between linguistic and cognitive deficits in AD, employing a variety of listening tasks used with the AD population, and 2) to provide clarification of the theorized dissociation of language processes (semantic, lexical, syntactic and phonological) in the communication deficits of AD. Specific aims of the project are to use the gating task, introduced by Gosjean (1), to study the effects of acoustic-phonetic and semantic cues on the process of spoken word recognition in normal young, normal elderly and AD subjects. In the gating task, subjects hear increasingly larger acoustic-phonetic fragments of a word over a series of trials and try to make an identification response. Subjects will hear words in isolation (Exp 1) or in the presence of semantically related or unrelated primes (Exp 2). The task simulates the acoustic input available to a listener in real-time and allows measurement of the amount of input necessary for word recognition under changing conditions. Results will provide basic information regarding the preservation of auditory comprehension and the nature of the semantic memory deficit in AD.